The AI revolution is here, you can’t help but read about it every day. In this series, we will talk about the different ways AI will affect testing and discuss whether testers will be safe from the changes that
AI will introduce into our development cycle.
In the near term (10 years maybe) I see the following being the major impacts AI will have on testing:
- Smarter test automation – Automation that evolves as the application changes.
- Bots for testing – Bots that crawl your application and try to detect changes and defects.
- Prioritized testing – Predicting where bugs have been introduced into the application.
- Testing the AI itself.
Interestingly the advent of mainstream AI and it’s rapidly increasing
rate of adoption may cause the first big shift within software testing. Software testing is already quite complex, but testing AI will only increase the complexity. Additionally, a program that uses AI will be more difficult to test as the program will no longer be deterministic.
Testers may need to pick up additional technical skills to better understand, and thus test programs utilizing AI. Testers will need to have a thorough understanding of how AI works in order to determine if it is working correctly.
Keep in mind that it is still possible to test AI. With knowledge of how the AI learns testers could put together
data sets which when consumed by the AI could produce known outcomes, i.e. if testing google search you would always expect IMDB to come up on the first page of results based on the current internet data set, if this wasn’t to happen then this would be an obvious bug.
So along with understanding AI testers will definitely need to understand data sets and how to manipulate them in order to create new test scenarios. Interestingly our current Test Case management tools may need to be improved in order to have tests linked to data sets and have the data sets version controlled.
That’s it for the first part of this series, I hope you’ve enjoyed it. Part two should follow soon.